ProRail is responsible for the maintenance, extension, management, and safety of the national railway network in the Netherlands. Therefore, it is important for ProRail to get a good picture of the condition of the various assets, so that they can be better maintained and timely replaced. This helps them to keep the cost down while ensuring safety.
Project
Project Insight and Forecasting
Project Insight and Forecasting
Client
ProRail
ProRail
Industry
Rail, Asset management
Rail, Asset management
Project type
Predictive Analytics, Asset Management
Predictive Analytics, Asset Management
Geography
Netherlands
Netherlands
Year
2020-current
2020-current
Website
prorail.nl
prorail.nl
Challenge
ProRail, the Dutch national railway infrastructure manager, faces thousands of decisions each year on when to maintain or replace critical assets. Traditionally, these decisions relied on inspections, experience, and rules of thumb — risking either premature maintenance (costly) or delayed intervention (unsafe).
Adding to the complexity: replacements and maintenance are financed by the government, requiring ProRail to forecast asset replacements up to 15 years in advance. Inaccurate lifespan predictions can lead to significant budget discrepancies. ProRail needed a data-driven approach to improve these forecasts across the entire Dutch rail network.
Our solution
Lynxx established the Asset Degradation Team (ADLT) within ProRail, developing a unified, data-driven methodology for predicting the remaining lifespan of railway assets. By analysing multi-year measurement data, we identify degradation trends and forecast when assets will require maintenance or replacement.
Our approach transforms raw measurement data into actionable maintenance forecasts. For track geometry, we process height deviation measurements from inspection trains, converting them into a Track Quality Index (TQI) per 200-metre section. By tracking TQI trends over multiple years, we generate reliable predictions of future track condition.
The methodology has been applied across multiple asset types, including:
- Track geometry — predicting when sections require tamping or realignment
- Rails — forecasting replacement needs based on wear patterns
- Overhead contact wires — predicting end-of-life for catenary systems
Results
For track geometry the Asset Degradation Team now delivers operational forecasts covering over 5,000 km of track. Processing 4.5 gigabytes of measurement data from 2016–2025 takes just 15 minutes, producing both current status and future projections. Three-year forecasts achieve 95% accuracy within a 0–28% error margin.
These predictions are visualised in interactive ArcGIS maps, enabling asset managers to generate maintenance plans directly from the data. The result: better planning, more efficient resource allocation, and ultimately a safer, more reliable railway, while supporting transparent multi-year budgeting for the Dutch government.
